Firstly, I would like to thank specially here both Adam and Guillaume, as this would not be possible without their great projects such as audioformats and arsd-official:image_files and arsd-official:ttf, but this project still had many discord contributors for my unending questions, I started using D before understanding anything about compilers, linkers or whatever, and my engine even has its almost complete standard library!.

This is the first time my engine is able to play sounds (took me a lot of effort to completely ditch SDL), but I would like to say that using SDL increases a lot of complexity in your project unnecessarily, you need so many old C dependencies (DLLs), which you have then to start maintaining all those DLLs to build to your target platform, and I believe a lot of people here knows how hard is to maintain C build systems specially when you're dealing with platforms besides Linux.

By completely ditching SDL, I was able to reduce my dependencies on Android from 200MB to 25MB. In the case of Xbox, it went from 40mb to 4mb (and at that time, image and audio loading weren't even working).

Congrats too for Walter developing the x64 output for DMD and kinke for LDC (LDC started distributing android compilers about the time I really joined D community).

Although many people here still says that D is dying, I can see how much of a milestone is doing that project, which has took me years in learning, testing, refactoring.

As this milestone was reached, I would like people to try coding D more, because, if I completely depended on C libraries to build my engine, I would be doomed! By the way, after doing bindings to Java, Lua, C++ and C in D, I can say that them are a lot easier to maintain than doing on those really old languages.

Here's a sample screenshot from it running on xbox

As you can see, there's still a lot to do on porting etc.

And I would like to say that I'm not stopping there, I still got tons of work to do, this is my first time here posting explicitly about it, because it has finally reached an usable state ( think of it currently as a native Love2D ), and, in no time, it will have many game features to make it a lot more "Engine" like.

If you wish to take a look into the current code development, here it is:
https://github.com/MrcSnm/HipremeEngine

On Tuesday, 15 November 2022 at 18:42:15 UTC, Hipreme wrote:

Congrats! It was impressing to see you advance that quickly on difficult ground.

On Tuesday, 15 November 2022 at 18:42:15 UTC, Hipreme wrote:

Nice, great to see other games and engines being developed in D. Does the game itself have a project page or is it mainly just for testing/prototyping with your engine?

On Wednesday, 16 November 2022 at 16:44:18 UTC, Kenny Shields wrote:

Yes! I'll share the code in the future on a separate repository as sample games that you can do in my engine. Next game coming up is a super mario bros clone, after I finish fixing it for Android (which is almost done)

On Tuesday, 15 November 2022 at 18:42:15 UTC, Hipreme wrote:

Can we use it on Windows?

On Thursday, 17 November 2022 at 01:20:59 UTC, zjh wrote:

Yes, HipremeEngine is being developed Windows first. Which just made me remember that I need to test on linux again.

On Thursday, 17 November 2022 at 11:21:35 UTC, Hipreme wrote:

If Linux will be supported it could be used on Steam Deck :)

On Thursday, 17 November 2022 at 12:02:07 UTC, Sergey wrote:

Linux is supported at the moment, it is just that I have been quite some time without testing, but there should not be too many difference. Android is linux too, and my engine does supports Android :D

Big hats off to you.

How hard was it in terms of general design, and how much math did you need to learn out of interest?

On Monday, 21 November 2022 at 01:14:41 UTC, thebluepandabear wrote:

The math is not that hard, as most of it depends on understanding specially the Model View Projection matrix, which you can find nice tutorials on internet. The rotation being the hardest part, although I did not use yet quaternions for that.

The engine design I was quite experienced in abstraction development, though, almost for every platform I added, I needed to refactor a bit, the progress is fairly incremental, I really doubt one could write a design for it and after adding a couple of platforms and libraries, it being the same way as in the start. One must always remember that for an abstraction, you need to use the most complex/verbose as the rule, the simple/direct one will be basically automatic to implement.

I believe the only code I yet need to touch again is the ErrorHandler, which I have it a bit underused as most of things just worked.